The Influence of Individualism, Collectivism, and Locus of

Control on Environmental Beliefs and Behavior

John A. McCarty and LJ. Shrum
This study investigates the influence of value orientations measured at the individual level
(individualism, collectivism, and locus of control) and of economic status on environmental
beliefs and behavior. Structural equation modeling reveals that the preferred model is one
in which the value orientations and economic status influence beliefs about recycling, which
in turn influence recycling behavior, but the influence of the value orientations and eco-
nomic status differs as a function of the type of environmental belief. Individualism and eco-
nomic status are related to beliefs about the inconvenience of recycling; collectivism and
locus of control are related to beliefs about the importance of recycling. Moreover, specific
beliefs about the importance of recycling have both a direct influence on recycling behavior
and an effect that is mediated by beliefs about the inconvenience of recycling. The authors
present the implications of the study for public policy and marketing communication efforts.

A s the environment has become an increasingly visi-

ble social and political concern in the last 20 years,
issues in the environmental domain have begun to
attract the attention of researchers in marketing and the
social sciences (e.g., Alwitt and Pitts 1996; Berger 1997;
Berger and Corbin 1992; Pieters et al. 1998; Shrum,
McCarty, and Lowrey 1995). Examples of such issues
include understanding consumer motivations underlying the
purchase of environmentally friendly products and services
(e.g., Bagozzi and Dabholkar 1994), explicating the relation
between general psychological constructs and environmen-
tal behavior (e.g., values, attitudes, beliefs; Berger and
Corbin 1992; Biswas et al. 2000; Dietz, Stern, and Guag-
nano 1998), and pinpointing the antecedents of post-
pure hase/postcon sumption behavior {e.g., household waste
disposal, product reuse; Alwitt and Pitts 1996; Shrum,
McCarty, and Lowrey 1995).
For the most part, understanding and predicting proenvi-
ronmental behavior has proved to be remarkably difficult.
One of the constant refrains in both the academic and popu-
lar press is that even though it seems as if nearly everyone
has positive attitudes and beliefs regarding the environment,
proenvironmental behavior (purchase, disposal) has not
been correspondingly ubiquitous. Public opinion polls con-
sistently fmd that a vast majority of people say that their
purchases are infiuenced by environmental concerns (Chase
and Smith 1992; Dagnoli 1991), indicate that they would
pay more for environmentally friendly products (Dagnoli
1991; Hume and Strnad 1989), and classify themselves as
environmentalists (Gutfeld 1991). Yet few "green" products
have been particularly successful (Reitman 1992), and
increases in recycling over the years have not been impres-
sive (Porter, Leeming, and Dwyer 1995).
JOHN A. MCCARTY is an assistant professor. School of Manage-
ment, George Mason University. LJ. SHRUM is an associate pro-
fessor, Department of Marketing, Rutgers University.
We propose that there are at least two interrelated reasons
it has been so difficult to understand and predict pro-
environmental behavior. First, the nature of the behavior
and the motivations underlying it are quite different from
most types of consumer purchase behavior. Most models of
purchase intention assume that the consumer has some level
of self-interest and that the purchase of a particular product
or service is driven by an assessment of the benefits that
would accrue directly to the individual or household, rela-
tive to the assessment of costs. Unlike most consumer
behaviors, however, the direct, extrinsic benefits that accrue
from proenviron mental behaviors (e.g., a noticeably better
environment) are at best long-term for the individual and
may in fact never be realized.' The costs, conversely, tend
to be realized immediately. Products thai are made from
recycled material may be more expensive or perceived as
inferior, and recycling requires certain behaviors that may
be perceived as inconvenient. Therefore, although there are
benefits and costs associated with proenviron mental behav-
iors, the nature of the benefits and the assessment of them
relative to the costs is likely a different matter than in the
case of most consumption activities (Thogersen 1996).
Second, because the nature of the benefits of proenviron-
mental behaviors and their assessment relative to their costs
may be different from those of other consumer behaviors,
the antecedents of proenvironmental beliefs may differ from
the antecedents commonly investigated by consumer
researchers. We propose that fundamental beliefs people
hold that pertain to their interaction with the world around
them and with other people influence the formation of their
beliefs about environmental issues and their propensity to
'Monetary incentives for engaging in recycling behaviors would be
direct and immediate benefits to individuals for engaging in the behavior.
Biswas and colleagues (2000) note, however, that [he use of economic
incentives, such as high redemption values and coupons, has decreased in
recent years because of the reduction in value of many recyclable materi-
als. Thus, direcl and immediate monetary incentives to individuals for
engaging in recycling behavior have declined as recycling programs have
developed.

Vol. 20(1)
Spring 2001, 93-104 Joumal of Public Policy & Marketing 93
94 Environmental Beliefs and Behavior
engage in proenvironmental behaviors. In particular,
because the benefits that accrue from proenvironmental
behaviors are future oriented and unlikely to benefit directly
the person performing the behavior, it is likely that funda-
mental concepts that relate to people's beliefs about their
ability to influence future outcomes and their desire to pro-
vide benefits for others may influence proenvironmental
beliefs and behaviors.
In this article, we develop and test a model of the rela-
tionship between such fundamental beliefs (or value orien-
tations) and proenvironmental beliefs and behaviors. We
propose that three basic values that relate to people's inter-
action with the world and with others will influence pro-
environmental behavior. Two of these constructs, individu-
alism and collectivism, relate to basic beliefs about humans'
relationships and interactions with other humans. The third
construct, locus of control, is a basic belief about humans'
relationships and interactions with their environment. We
test this model for a particular proenvironmental behavior
(recycling) and beliefs related to this behavior. Because
other factors (e.g., socioeconomic variables) have been
shown to relate to recycling (Berger 1997), we test the fun-
damental beliefs of interest in conjunction with the variable
of economic status.
Background
Fundamental Beliefs: Value Orientations
Individualism, collectivism, and locus of control are basic
beliefs that people hold with respect to their interaction with
others and the world around them. In that sense, these basic
beliefs can be referred to as value orientations (Kluckhohn
1951), which have been described as fundamental beliefs
that cultural groups or individuals have that assist in the
adaptation to their physical and social environment.
Although these value orientations have often been investi-
gated by social scientists at the cultural level (e.g., differ-
ences in individualism across countries), the current investi-
gation focuses on these beliefs at the individual,
psychological level.
Individualism and Collectivism
Individualism and collectivism have been identified as ori-
entations that can be taken with respect to a person's or
group's relationship to others. Individualism can be broadly
characterized as the tendency to value the individual over
the group and give priority to persona! goals over group
goals (Triandis 1989). Individualism stresses individual ini-
tiative, a greater focus on the self, and emotional indepen-
dence (Hofstede 1980). It also emphasizes self-reliance and
freedom of choice (Bellah et al. 1985), stresses individual
rights over duties, and puts emphasis on cost-benefit analy-
ses in determining behavior (Triandis 1994). In contrast,
collectivism emphasizes the goals of the group over per-
sonal goals, stresses conformity and in-group harmony, and
defines the self in relation to the group (Triandis 1995). Col-
lectivism emphasizes sharing, duties, and obligations (Hof-
stede 1980).
When studied at tbe cultural level, individualism and col-
lectivism are considered to represent opposite ends of one
continuum, and cultures are often described as being either
individualistic or coUectivistic in their orientation (Triandis
1994). At the individual level, however, research suggests
that individualism and collectivism represent separate
dimensions '(Triandis and Gelfand 1998). Both individual-
ism and collectivism can exist within the same culture
(Sinha and Tripathi 1994; Triandis 1994), and a person may
possess both individualistic and coUectivistic tendencies
(Sinha and Tripathi 1994; Triandis 1989, 1994). Different
situations may cause a person to sample individualistic or
coUectivistic aspects of the self (Traflmow, Triandis, and
Goto 1991). Thus, a person may believe in personal initia-
tive and independence yet also value group harmony and
sharing. Because the present study involves individual-level
activity, we treat individualism and collectivism as two sep-
arate dimensions that pertain to the ways in which individu-
als relate to other individuals.
Locus of Control
Locus of control refers to the extent to which people believe
that they have the ability to affect outcomes through their
own actions (Rotter 1966). On the one hand, some people
believe that they have substantial infiuence over their lives
and that their actions infiuence particular outcomes. These
people are said to have an internal locus of control (inter-
nals). They typically perceive themselves to have control
over their future and believe that outcomes are related to the
work they put into them (Lefcourt 1991). On the other hand,
people who believe that they are relatively powerless and
have little infiuence over outcomes are said to have an exter-
nal locus of control (externals). Triandis (1984) has noted
that this psychological construct of locus of control is equiv-
alent to the cultural-level value orientation dimension of
beliefs about human-nature interactions, which relates to
the extent to which a cultural group believes that it is supe-
rior to nature, lives in harmony with nature, or is subjugated
to nature.
Beliefs Related to Recycling
Although we expect that these abstract, fundamental value
orientations (individualism, collectivism, and locus of con-
trol) will relate to people's propensity to engage in recy-
cling, we do not expect that their effects on behavior will be
direct. Instead, we expect that these fundamental values will
infiuence more specific psychological constructs (beliefs
about recycling), which in turn will infiuence behavior.
Research in a variety of areas has shown that specific atti-
tudes and beliefs often mediate the values-behavior relation
(e.g., Alwitt and Pitts 1996; Homer and Kahle 1988;
McCarty and Shrum 1994).
Two particular belief constructs that have consistently
been shown to relate to recycling are a general attitude or
belief about the importance of recycling and a specific belief
about the inconvenience of recycling (Hines, Hungerford,
and Tomera 1987; McCarty and Shrum 1994; Vining and
Ebreo 1990). Therefore, we consider these two belief con-
structs as ones that potentially mediate the relations between
value orientations and recycling behaviors.
Importance and Inconvenience can be contrasted on three
important and related dimensions that pertain to how they
potentially mediate relations between the abstract, funda-
mental beliefs and specific behaviors. First, the importance
 Journal of Public Policy & Marketing
of recycling relates to the benefits of engaging in the behav-
ior (e.g., a cleaner environment), whereas inconvenience
focuses on the costs (e.g., the time required to prepare mate-
rials for collection). Second, the importance of recycling is
a long-term consideration, but inconvenience is primarily
short-term in nature. As we have noted, the benefits of the
environment will generally only be realized in the future,
and therefore people may not see these positive results in
any immediate way, whereas the costs to the individual in
terms of engaging in the behavior are relatively immediate.
Third, the two belief constructs differ in their level of
abstraction. Because the importance of recycling relates to
long-term rewards for the environment and society, these
beliefs tend to be more abstract and general in nature than
beliefs about the inconvenience of recycling, which focus on
immediate costs to the individual and tend to be more con-
crete and specific.
Economic Status
Although the individual-level values and beliefs we have
described are the main focus of the study, other variables
have been shown to be related to recycling. Several studies
have investigated the relationship of demographic variables
and recycling behavior, showing mixed results with respect
to the impact of these variables on environmental behaviors
(for reviews, see Berger 1997; Shrum, Lowrey, and
McCarly 1994). Berger (1997) argues that the role of
socioeconomic factors such as income and education may
be a complex one, and otber factors may mediate the rela-
tionship between these variables and behavior. For example,
socioeconomic factors may facilitate conditions for recy-
cling; Wealthier neighborhoods may have more access to
recycling facilities, and a household's recycling may there-
fore be as much a function of the economic conditions of the
community as of the household's own income level. A study
by Jacobs, Bailey, and Crews (1984) supports this notion,
fmding that participation in recycling was related to the
housing prices of a neighborhood. Along these same lines,
Berger also notes that perceptions of the inconvenience of
recycling may vary as a function of socioeconomic status: It
may simply be more inconvenient for people living in
smaller houses to store recyclable materials until the day
they can be left for curbside pickup.
Proposed Model
The research just reviewed provides information that is use-
ful in developing specific hypotheses about how the con-
structs of individualism, collectivism, and locus of con-
trolall considered at the psychological levelas well as
the construct of economic status might influence recycling
behavior through the recycling beliefs about importance and
inconvenience.
Individualism, Collectivism, and Recycling
The nature of the two constructs, individualism and collec-
tivism, suggests that tbey may relate in different ways to the
propensity for people to engage in proenvironmental behav-
iors. On the one hand, Sinha and Verma (1987, p. 124) note
that individualism promotes "contractual relationships
which are based on the principles of exchange. People cal-
culate profit and loss before engaging in a behavior." Indi-
95
vidualistic people also place greater importance on the rela-
tion between their behavior and their own needs and beliefs
(Leung and Bond 1984). On the other hand, collectivism
tends to promote a consideration of the implications of peo-
ple's behavior for others. The behavior of people who are
collectivistic tends to be driven by social norms, and these
people show a willingness to share scarce resources (Sinha
and Verma 1987). Thus, it is reasonable that a behavior such
as recycling may be a function of individualism and collec-
tivism. Individualism tends to focus a person's attention
toward immediate benefits relative to costs. In the short run,
however, extrinsic benefits are few relative to the perceived
costs (e.g., inconvenience). Collectivist tendencies should
focus a person toward group benefits, even if these benefits
are not immediate.
There is some limited empirical evidence for these con-
jectures. McCarty and Shrum (1994), using a student sam-
ple, found that collectivism is positively related to the level
of reported recycling behavior, and Hopper and Nielsen
(1991) provided evidence that participation in a community
recycling program is influenced by social norms. Dunlap
and Van Liere (1984) found that traditional American (indi-
vidualistic) values are negatively related to beliefs about
resource conservation.
Therefore, people who are more individualistic will tend
to focus on personal goals, and their behavior will tend to be
guided by immediate benefits relative to costs. Such people
are likely to consider recycling costly and as having few
immediate benefits, and they are unlikely to perceive recy-
cling as contributing to tbeir personal goals. Given this, we
predict the following:
H|:The degree of individualism will relate positively to the
beliefs about the inconvenience of recycling. People who are
more individualistic will tend to believe that recycling is more
inconvenient than will those who are less individualistic.
People who are more collectivistic are more concerned
with the good of the group and focus on group goals, shar-
ing, and duty. Because collectivism relates to the impact of
actions on the collective (society) and on the future, people
who are more collectivistic should consider recycling more
important than will those who are less collectivistic. There-
fore, we predict the following:
H2: The degree of collectivism will relate positively to the
beliefs about the importance of recycling. People who are
more collectivistic will tend to believe that recycling is
more important than will those who are less collectivistic.
Implicit in these two hypotheses is the assumption that
individualism and collectivism are independent constructs
rather than opposite ends of a single continuum, consistent
with previous research on these constructs at the psycholog-
ical level.
Locus of Control and Recycling
Several studies have shown that locus of control (or variables
similar to it) is associated with environmental issues. Balder-
jahn (1988) found that greater perceived ideological control
is positively related to attitudes toward ecologically con-
scious living and environmentally friendly purchase behav-
ior, and Schwepker and Cornwell (1991) reported evidence
showing that an internal locus of control is related to the
96 Environmental Beliefs and Behavior
propensity to purchase ecologically packaged products. In a
meta-analysis of previous work on environmental behavior,
Hines, Hungerford, and Tomera (1987) concluded that an
internal locus of control is positively related to environmen-
tally responsible behavior. As we have noted, many polls
show that most people have a positive attitude toward the
environment. An internal locus of control might provide peo-
ple with the belief that they can do something to change the
environment; thus, internals should be more likely to believe
that their actions with respect to the environment are worth-
while. Therefore, we predict the following:
H3: An internal locus of control will relate positively to the beliefs
about the importance of recycling. People who are more inter-
nal will tend to believe that recycling is more important than
will those who are less intemal (more extemal).
Economic Status and Recycling
Given previous research that suggests a relationship of the
economic status of people and their communities with recy-
cling, we expect that the income of a family, as well as the
economic level of a community, will relate to perceptions of
the inconvenience of recycling, thereby indirectly influenc-
ing the likelihood that a person will engage in recycling
behavior. Specifically, we predict tbe following:
H4: Economic status will he negatively related to perceptions of
the inconvenience of recycling. People with lower incomes
and/or living in areas of lower econorfiic status will believe
recycling is more inconvenient than will those with higher
incomes and/or living in areas of higher economic status.
Beliefs About Recycling and Behaviors
The differences in the nature of beliefs about the importance
and inconvenience of recycling lead to expectations about
how these beliefs might relate to recycling behaviors. Given
the differences in the level of abstraction and specificity of
importance and inconvenience, we expect that the more gen-
eral beliefs about importance will influence the more specific
beliefs about inconvenience. That is, as people develop more
positive beliefs about the long-term importance of recycling,
they should hold less negative beliefs about the inconve-
nience of engaging in the behavior. Therefore, although we
expect that beliefs about the importance of recycling will
relate directly to the likelihood that a person will engage in
recycling behaviors, these general beliefs about importance
should also influence perceptions of the inconvenience of
engaging in the bebaviors, which in turn will influence die
behavior. Therefore, we predict the following:
H5: Beliefs about the importance of recycling will have a direct,
positive relationship with recycling behaviors. People who
believe that recycling is more important will tend to report
that they recycle more than will those who believe that recy-
cling is less important.
Hg: Beliefs about the importance of recycling will be negatively
related to beliefs about the inconvenience of recycling. Peo-
ple who believe that recycling is more important will tend to
believe that recycling is less inconvenient than will those
who believe that recycling is less important.
H7: Beliefs about the inconvenience of recycling will have a
direct, negative relationship with recycling behaviors. Peo-
ple who believe that recycling is more inconvenient will
tend to report that they recycle less than will those who
believe recycling is less inconvenient.
Figure 1 shows the hypothesized relationships of the pro-
posed model. We expect that the influence of the value ori-
entations and economic status (the four exogenous vari-
ables) on recycling behavior will be mediated by recycling
beliefs (importance and inconvenience) and in the specific
ways shown in the figure. We do not expect otber relation-
ships between the exogenous variables and the mediated
variables to be present. Thus, individualism and economic
status are not expected to affect the importance of recycling,
and collectivism and locus of control are not expected to
relate directly to the inconvenience of recycling.
Method
Sample and Procedure
The data for the study were collected through a mail ques-
tionnaire sent to a stratified random sample of adults in a
heavily populated Midwestern state. Only people who lived
in communities in which curbside recycling was available
were included in the sampling frame. Therefore, cities and
suburbs in the state were first selected on the basis of the
availability of recycling. Individuals within these strata
were then randomly selected for inclusion in the study.
Because of the need to restrict sampling to strata that pro-
vided recycling opportunities, neitber the most rural areas of
the state nor the major city in the state was represented in the
sample. In general, the sample was obtained from ZIP codes
in cities of 10,000 to 250,000 people and from ZIP codes of
the suburbs of the major urban areas.
Each member of the sample was sent a questionnaire with
a cover letter explaining that the study was designed to
investigate the beliefs and activities of the people living in
the state. The questionnaire included items on a variety of
topics, including those of interest for the curTent study.
Therefore, although several questions pertained to recy-
cling, the focus of the study was masked to a large degree by
the inclusion of items unrelated to environmental issues.
Following the recommendations of Dillman (1978),
reminder postcards were sent out approximately three weeks
after the original mailing to people who had not returned the
questionnaire. Members of the sample who had not returned
the questionnaire after another two weeks were sent a new
cover letter and replacement questionnaire.
A total of 1891 surveys were mailed; 30 were returned as
undeliverable, which left 1861 presumably delivered. Of the
1861 surveys delivered, 727 were returned by the respon-
dents, for a return rate of 39.1%. Of those responding,
57.9% were male,2 the median income was $45,000, age
ranged from 18-92 years with an average of 48 years, and
the average number of years of education was 14.6. A total
of 193 respondents were excluded from the analyses
because they had missing data on one or more measures
used in the present study. Thus, the analyses were based on
534 respondents.
^The overrepresentation of male respondents was a function of our selec-
tion criteria and not a result of response bias. Parts of tbe questionnaire
were used for purposes unrclaied to this study, and Ihese purposes necessi-
tated holding sex constant for a subset of the entire sampling frame.
 Journal of Public Policy & Marketing
Figure 1. Proposed Model of the Relations Among Value Orientations, Economic Status, Recycling Beliefs, and Behavior
Measures
We measured each latent construct using multiple indica-
tors. The particular items used to measure each construct
appear in Table 1.
Exogenous Measures
The four exogenous latent factors in the study were eco-
nomic status, individualism, collectivism, and locus of con-
trol. Three variables were used to measure economic status.
One of these was the family income of the respondent as
reported in the questionnaire. Respondents selected 1 of 11
categories; the first was $0-$10,000, and the categories con-
tinued with $iO,000 ranges up to $100,000. The final cate-
gory was "more than $100,000." The other two variables
considered for the economic status construct were from the
1990 U.S. Census. These were the median housing value
and the per capita income of the ZIP code area in which each
respondent resided.
Five items were used to measure collectivism and three
items were used to measure individualism. These items were
similar to ones used in previous studies (Breckler. Green-
wald, and Wiggins 1986; Han 1991). Respondents were
asked to evaluate these items on five-point scales with
anchors of "not at all important" (1) and "extremely impor-
tant" (5).
Ten items from a scale developed by Paulhus (1983) were
used to measure iocus of control. Paulhus partitioned locus
of control into three distinct and independent behavioral
spheres; the sphere deemed most appropriate for the current
97
study was sociopolitical control (control in large-group
[societal] situations). The items composing this measure of
locus of control scale were evaluated on six-point scales
with anchors of "strongly disagree" (1) and "strongly agree"
(6). Higher scores on these measures indicate a more inter-
nal locus of control.
Endogenous Measures
The measures of beliefs ahout the inconvenience and impor-
tance of recycling were also evaluated on six-point agree-
ment scales and were similar to ones used in previous stud-
ies (McCarty and Shrum 1994; Williams 1991). Two items
were related to the inconvenience of recycling and three
items were related to the importance of recycling.
The behavioral measures consisted of three items that
measured the extent to which respondents recycled newspa-
pers, cans, and glass/jars in their households. The frequency
of these recycling activities was evaluated on seven-point
scales with anchors of "never" (I) and "always" (7).
Analysis
We used structural equation analysis by means of EQS
(Bentler 1995) to evaluate a measurement model and a
causal model simultaneously, using the maximum likelihood
estimation method. Thus, the analysis evaluated the extent to
which the observed variables were indicators of the latent
constructs in the model and the strength of the relationships
among the latent variables as specified by the hypothesized
paths. We evaluated the viability of the specific hypotheses
98 Environmental Beliefs and Behavior

Table 1. Measurement Model Results

Standardized
Factor
Construct Loading t-Value
Economic Status (a = .761)
Income .377 8.45
Median household value (of ZIP code area) .932 23.37
Per capita income (of ZIP code area) .912 11.21
Individualism (a = .440)
Being unique, different from others in many respects. .418 4.93
Being competitive with others. .276 3.74
Working independently from others. .698 5.43
Collectivism (a = . 746)
Working hard for the goals of a group even U doesn't result in personal recognition. .555 12.42
Being a cooperative participant in group activities. .570 11.24
Readily helping others in need of help. .647 13.58
Doing what is good for most of the people in the group, even if it means that the individual
will receive less. .607 13.27
Sharing with others. .661 13.56
Locus of Control (a = .816)
There is very little we, as consumers, can do to keep the cost of living from going higher. (R) .521 12.59
With enough effort we can wipe out political corruption. .528 12.35
The average citizen can have an influence on government decisions. .730 20.57
This world is run by the few people in power, and there is not much the little guy can do about it. (R) .545 12.38
By taking an active part in political and social affairs we, the people, can control world events. .743 18.36
When I look at it carefully, I realize it is impossible to have any really important influence over
what politicians do. (R) v679 17.22
I prefer to concentrate my energy on other things rather than on solving the world's problems. (R) .328 7.00
One of the major reasons we have wars is because people don't take enough interest in politics. .372 8.28
In the long run we, the voters, are responsible forbad government on a national as well as local level. .512 11.48
It is difficult for people to have much control over the things politicians do in office. (R) .600 13.94
1 Recycling Beliefs: Importance (a = .643)
Recycling will reduce pollution. .608 a
Recycling is important to save natural resources. J52 7.63
Recycling will save land that would be used as dump sites. .674 7.87
2 Recycling Beliefs: Inconvenience (ex = .685)
Recycling cans, bottles, newspapers, etc. is inconvenient. .747 a
I hate having to wash out bottles for recycling. .690 . 10.16
j Recycling Behaviors (a = .874)
I recycle newspapers used at home. .818 a
I recycle cans used at home. .803 15.53
I recycle glass jars/bottles used at home. .889 20.12

Notes: (R) indicates items that werereversed-scored, a = fixed during analysis, no t-values given. All loadings are significant atp < ,001.

by the significance of specific paths between latent con-
structs. We assessed the reliabilities of the measures by
Cronbach's alphas. We tested individual parameters in the
proposed model (factor loadings and hypothesized paths of
the model) by univariate t-tests and used robust standard
errors (Bentler 1995) to form the test statistics because of
their ability to protect against violations of the assumption of
multivariate normality of the observed variables. We evalu-
ated the overall ftt by the standardized root mean squared
residual (SRMR; Bentler 1995) and the root mean square
error of approxitnation (RMSEA; Steiger 1990).^
difference in x^ values for nested models (as these models are) is
itself distributed as a x^ with degrees of freedom equal to the difference in
The overall fit of the proposed model was evaluated in a
nested manner as recommended by Anderson and Gerbing
(1988). The proposed model was tested against a baseline
model that included no directional paths among the latent
the degrees of freedom of the models being considered. The other two fit
statistics were selected because they are sensitive to different aspects of
evaluating a proposed model. The SRMR is an analysis of the residuals
between the hypothetical covariance matrix and the fitted covariance matrix;
Hu and Bentler (1998) indicate that anwng indices they tested, the SRMR
was the most sensitive to misspecified factor covariances. The RMSEA is an
indication of the lack of fit of the model to the population covariance matrix;
Hu and Bentler (1998) found that it is particularly sensitive to models with
misspecified factor loadings. Hu and Bentler (1999) suggest a cutoff of .08
for the SRMR iwid .06 for the RMSEA before researchers conclude that they
have an adequate fit of a hypothesized model.
 Journal of Public Policy & Marketing 99

Table 2. Causal Model Results: Correlations Among Exogenous Constructs and Standardized Path Coemdents Among Latent
Constructs

Proposed Model Mediated/Direct Model

Coefficient t-Value CoefTicient t-Value
Correlations Among Exogenous Variables
^12'. Economic status <- individualism .034 .54 .029 .51
013: Economic status <-> collectivism -.066 -1.24 -.066 -1.25
(fi4: Economic status <-> locus of control .009 .19 .009 .19
^23- Individualism <-> collectivism -.023 -.33 -.041 -.63
(t24: Individualism *-* locus of control -.076 -1.09 -.090 -1.39
^24- Collectivism <-> locus of control .252 4.62*** .252 4.63***

Standardized Path Coefficients

Y21: Economic status * inconvenience -.197 -4.58*** -.169 -3.74**
Y22- Individualism -* inconvenience .138 L96* .139 2.01*
713: Collectivism -^ importance .223 3.20** .225 3.23**
7)4: Locus of control > importance .307 4.16*** .304 4.11***

731: Economic status * recycling behaviors .115 2.64**

732: Individualism > recycling behaviors .051 .95
733: Collectivism - recycling behaviors -.007 -.12
734: Locus of control > recycling behaviors .021 M
p2i- Importance > inconvenience -.387 ^99*** -.386 ^99***
P31: Importance > recycling behaviors .139 2.12*** .148 1.87
p32- Inconvenience recycling behaviors -.544 -7.36*** -.516 -6.65***

R2 R2
(Importance) .179 .177
(Inconvenience) .207 .198
(Recycling behaviors) .375 .378

*p < .05.
*V< 01.
**V<.ooi.

constructs and a model that included the hypothesized
mediated paths as well as direct paths from the exogenous
constructs and recycling behavior (mediated/direct model).
The appropriateness of these models was evaluated hy the
change in x^ relative to the change in degrees of freedom.
The Satorra-Bentler scaled x~ (Bentler 1995) was used
because it is robust to deviations from multivariate normal-
ity. The Lagrange-Multiplier test (Bentler 1995) was used to
evaluate whether the addition of individual paths not
hypothesized in the proposed model would significantly
improve the fit of the model.
Results
Measurement Model
Table I provides information about the quality of measure-
ment of the latent constructs in the model. As the standard-
ized factor loadings in the table show, all the measured vari-
ables loaded on the latent constructs as expected. Except for
the individualism construct, the Cronbach's alphas for the
measured variables for each of the latent constructs were in
the acceptable range, which indicates that they are generally
reliable measures of the latent constructs. The low alpha for
individualism suggests that it was not measured reliably.
and therefore its relation with other variables should be
interpreted with caution.
An additional consideration of the measurement model
relates to our expectation that individualism and collectivism
would emerge as distinct constructs at the psychological
level As the top portion of Table 2 shows, the correlation
between the individualism and collectivism factors was not
significant for the proposed model. The only significant cor-
relation among the exogenous factors was a moderate, posi-
tive relation between collectivism and locus of control.
Therefore, the different exogenous variables can be consid-
ered to represent relatively independent sources of variation.
Proposed Model: Hypothesized Relationships
Exogenous Predictors
We expected that the degree of individualism would relate
positively to perceptions of the inconvenience of recycling;
people who are more individualistic should perceive recy-
cling as more inconvenient than will those who are less indi-
vidualistic. As the middle portion of the proposed model
columns of Table 2 shows, this path was significant and in
the predicted direction. Thus, H| was supported. We also
expected that degree of collectivism (H2) and intemal locus
100 Environmental Beliefs and Behavior
Table 3. Model Fit Summary for Competing Models
Degrees of
Model X^ Freedom
Baseline 941.86 372
Proposed 707.37 364
Mediated/direct 699.44 360
/<-01.
of control (H3) would relate positively to the importance of
recycling. The results shown in Table 2 (proposed model
columns) indicate that these hypotheses were supported.
The standardized path coefficients for each of these paths
were significant and in the predicted direction. Finally, we
expected that economic status would show a negative rela-
tionship with inconvenience such that people with a lower
economic status would perceive recycling as more inconve-
nient than would those with a higher economic status (H4).
As the results for the proposed model show, this hypothesis
was supported.
Endogenous Predictors
We expected that beliefs about the importance of recycling
would relate positively to the recycling behavior (H5) and
negatively to beliefs about the inconvenience of recycling
(H^). The results shown in the bottom portion of Table 2 for
the proposed model indicate that these two hypotheses were
supported. We also expected that beliefs about the inconve-
nience of recycling would relate negatively to recycling
behavior (H7); this too was supported. Thus, beliefs (both
importance and inconvenience) were related to recycling
behavior in the expected ways. As shown at the bottom of
Table 2, somewhat more than one-third of the variability m
recycling behaviors is explained by the proposed model
Mediated/Direct Model
The mediated/direct model of Table 2 provides a test of the
extent to which the relationships of the exogenous variables
(the fundamental values and economic status) to recycling
behavior are mediated by the recycling beliefs. Complete
mediation is evident if the direct paths between the exoge-
nous variables and recycling behavior are not significant.^
When direct relationships between the exogenous vari-
ables and recycling behavior were considered in conjunction
with mediated relationships, there was a significant, direct
relation between economic status and recycling behaviors,
as shown in Table 2. Economic status appears to have an
^Thc comparison of the proposed model with this mediated/direct model
represents an important consideration, because it is necessary for the medi-
ated paths to be significant in the presence of the direct paths from the
exogenous variables to behavior for mediation to be evident (Baron and
Kenny 19S6); a completely mediated relationship is suggested if the direct
path that corresponds to a mediated relationship is not significantly differ-
ent from zero when niediation is tested. According to Baron and Kenny,
additional requirements to establish mediiUion are [hat (I) the exogenous
variables are correlated with the outcome variables, (2) the exogenous vari-
ables are correlated with the mediators in the expected manner, and (3) the
mediator affects the outcome variable. Tests were conducted that deter-
mined that these three requirements were met.
ADegrees of
Freedom SRMR RMSEA
.092 .057
234.49* 8 .052 .046
7.93 4 .051 .046
influence on behavior that is mediated by inconvenience, as
hypothesized, and a direct effect that is over and above this
mediated influence. All other direct relationships between
the exogenous factors and recycling behavior were not sig-
nificant, which indicates that the relation of collectivism and
locus of control with recycling behavior was completely
mediated by importance and that the influence of individu-
alism on recycling was completely mediated by beliefs
about the inconvenience of recycling.
Tests of Additional (Nonhypothesized) Paths
We used the Lagrange-Multiplier test to determine whether
additional paths from the exogenous variables to the media-
tors (e.g., economic status to importance) would provide a
significant improvement in fit. No additional paths were
found to improve the fit of the model significantly. Thus, as
expected, degree of individualism and economic status were
unrelated to beliefs about the importance of recycling; col-
lectivism and locus of control did not show significant,
direct relationships with beliefs about the inconvenience of
recycling.
Hierarchical Tests of Models
Table 3 shows the measures of overall goodness of fit for a
null model (a model specifying no directional paths among
the latent constructs), the proposed model, and the medi-
ated/direct model. As the fit indices indicate, the proposed
model fit the data wetl in that the SRMR was below .08 and
the RMSEA was below .06. Moreover, as the change in x^
relative to the change in degrees of freedom indicates, the fit
of the proposed model was a significant improvement over
the fit of the null model.^ The change in x^ relative to the
change in degrees of freedom for the mediated/direct model
compared with the proposed model suggests that estimating
all of the direct relationship between the exogenous con-
structs and recycling behavior did not significantly improve
the fit.6
'The removal of the directional paths between the latent variables to test
the baseline model led to the underidentification of the etidogenous vari-
able inconvenience because there were only two indicators of this latent
constnict. The error variances of the two indicators of this construct were
constrained to be equal, which thus corrected the underidentification for
this baseline model.
^A model, similar to the proposed model but with the addition of the
direct path of economic status to recycling behaviors, significantly
improved the fit compared with the proposed nfwdel. This would be
expected, because this direct path was significant for the niediated/direct
model. Al! the significant hypothesized paths of the proposed model were
also significant for this revised model.
 Journal of Public Policy & Marketing
Discussion
The results of this study provide new infortnation regarding
the antecedents of recycling behavior. Previous research has
shown, for example, that beliefs about the importance of
recycling and beliefs about the inconvenience of recycling
are stable predictors of the propensity of consumers to recy-
cle. These findings are intuitive. However, this study's
results suggest that these relationships are not that simple.
For some people, such as those who are more individualis-
tic or have a lower economic status, the importance of recy-
cling is not a motivating issue; only inconvenience predicts
recycling behavior. For others, such as those who are more
collectivistic or have a more internal locus of control,
beliefs about the importance of recycling are positively
related to the propensity to recycle. As are the constructs of
individualism and economic status, collectivism and locus
of control are related to beliefs about the inconvenience of
recycling, but in a different way: Collectivism and locus of
control relate to beliefs about inconvenience, but only
through the mediating influence of beliefs about the impor-
tance of recycling.
The results of this study also shed some light on the rela-
tion between economic status and recycling behavior. Some
research suggests that this relation is mediated by access to
recycling {Berger 1997). That is, people who have a higher
economic status tend to have greater access to recycling
than do those who have a lower economic status, and this
access differential influences recycling behavior. Implicit in
this notion is that less access leads to more inconvenience.
The results of this study support this contention: Economic
status was negatively related to beliefs about the inconve-
nience of recycling, which in turn negatively influenced
recycling behavior. Note, however, that economic status did
not show any relation lo beliefs about the importance of
recycling, only to inconvenience.
Implications for Marketing and Public Policy
The importance of these findings to marketing and public
policy practices centers on the complex interplay of the
antecedents of recycling behavior. This study focused on
two belief constructs that have consistently been shown to
predict recycling behavior: inconvenience and importance.
The results of this study depart from previous ones by
demonstrating that the manner in which beliefs about the
importance and inconvenience of recycling are predictive of
behavior varies across types of people.
We have shown that importance has both a direct effect
on recycling behavior and an effect on perceptions of the
inconvenience of recycling, which in turn predict recycling
behavior. To the extent that people's perceptions about the
importance of recycling can be changed, this should have a
two-pronged influence on behavior. First, as people's
beliefs in the importance of recycling increase, the likeli-
hood that they will engage in the behavior should also
increase. Second, increasing people's beliefs in the impor-
tance of recycling should, over time, change their percep-
tions of the inconvenience of recycling. That is, as con-
sumers begin to understand the importance of recycling for
the environment and for their own and future generations,
this should influence their thoughts about the inconvenience
101
of engaging in recycling behaviors. Thus, importance will
have an influence on recycling behavior through this indi-
rect route.
The extent of such efforts, however, will likely depend on
the value orientations held by particular people. Our main
focus in this study was how these value orientations relate to
recycling beliefs and behavior, and our findings with respect
to these relations also have certain policy implications. The
implications are mainly related to segmentation and target-
ing efforts and in particular to what types of messages are
constructed to persuade people to recycle more. One clear
implication is that because the path to behavior varies as a
function of value orientation, the appeal of a persuasive
communication should also vary according to the value ori-
entation of the target. In particular, the importance of recy-
cling is a critical mediator of recycling behavior and thus
should be addressed in a persuasive communicationbut
only for people high in collectivism. For those high in indi-
vidualism, only inconvenience is a potential mediator, and
therefore attempts to change beliefs regarding the impor-
tance of recycling for those people may have little direct
effect on recycling behavior.
These recommendations are consistent with several stud-
ies in cross-cultural consumer behavior. For example, the
persuasiveness of advertising appeals has been shown to
vary across cultures; appeals that stress individualistic ben-
efits are more persuasive in individualistic cultures than col-
lectivistic cultures {United States versus Korea), and appeals
that stress group {family) benefits are more persuasive in
collectivistic cultures than in individualistic cultures {Han
and Shavitt 1994). Therefore, communications aimed at
changing recycling attitudes and behavior should attempt to
match the focus of the benefits (individual/inconvenience
versus group/importance) to the predominant value orienta-
tion of the target. Other research provides more support for
this notion: Aaker and Maheswaran {1997) found that the
extent to which people consider particular cues or informa-
tion (e.g., consensus information versus attribute informa-
tion) pertinent to a particular judgment varies across cul-
tures. Finally, research has also shown that the
persuasiveness of types of emotional appeals {e.g.. ego-
focused versus other-focused) used in persuasive communi-
cations varies as a function of the value orientation of the
message recipient {Aaker and Williams 1998). The results of
these studies, coupled with the results of our study, provide
clear direction in the design of persuasive communications.
In terms of segmentation, the results of this study suggest
that the easiest people to infiuence may be those who are
collectivistic and/or have an internal locus of control. Com-
munication efforts need to remind people high in collec-
tivism of the importance of the behavior for the society.
People with an intemal locus of control may only need to be
reminded that their behavior in the area of the environment
will make a difference.
Others may require more effort to reach and convince.
People high in individualism but low in collectivism are
probably the most difficult to change. If people who are high
in individualism tend to think in terms of short-term benefits
relative to costs, marketing communications efforts may be
used to shift the time frame and context of the benefits that
recycling provides to coincide with the short-term nature of
102 Environmental Beliefs and Behavior
perceived costs. That is, communication efforts can focus on
how the benefits of recycling are more immediate and can
hit closer to home than people might think. There is some
evidence that such a strategy would pay off. White (1999)
reported that the city of Boston used an advertising cam-
paign to influence nonrecyclers that focused on neighbor-
hood-specific themes and consequently made the benefits
more immediate and localized to the nonrecyclers' areas.
Another approach that may work with people who are
high in individualism is to make other rewards and benefits
salient to them. De Young (1985-86) has shown that intrin-
sic motivations (i.e., personal satisfaction) can play a role in
recycling behavior. To the extent that these immediate
intrinsic rewards can be made salient to individualists, such
benefits may aid in overcoming these people's resistance to
recycling because of their typical perceptions of high costs
relative to low short-term benefits.
The most interesting communications challenge relates to
this study's finding that individualism and collectivism can
coexist within a person, which replicates previous work on
the psychological nature of these constructs. As noted, a
person can be high in both individualism and collectivism.
The marketing challenge in these cases may be to induce
people to think collectivistically rather than individualisti-
cally. As Trafimow, Triandis, and Goto (1991) argue, dif-
ferent situations and contexts ean influence whether people
sample either their collectivistic or individualistic selves.
This suggests that marketing communication efforts aimed
at increasing recycling can benefit from an approach that
causes people high in individualism to sample their collec-
tive selves rather than their individualistic selves, thus shift-
ing their considerations from the costs of the behaviors to
the rewards.
Although segmenting on the basis of such individual dif-
ferences has typically been viewed as too difficult, it may be
easier than it sounds. Given the fragmented and nanowly
targeted nature of today's media (Turow 1997), it may be
possible and even relatively easy to reach audiences that
hold similar basic beliefs. Thus, if the readership of a par-
ticular magazine is known to have a more individualistic or
collectivistic orientation (e.g., readers of Forbes versus
readers of The Progressive), different appeals can easily be
constructed for targets with different value orientations.
Although we have stressed the contribution of individual,
psychological constructs (i.e., value orientations) on recy-
cling behavior and their implications for policy, the rela-
tionship of economic status (primarily the economic status
of the community) and recycling bebavior demonstrates that
efforts to increase recycling compliance should be consid-
ered at various levels. Community-level action that will
make recycling more accessible, particularly in lower
income areas, will complement the segmentation and com-
munication efforts we have described.
Limitations of the Study
There are a few limitations to the present study tbat readers
should consider wben drawing conclusions from the results.
The first limitation pertains to the narrow focus of the study.
Our primary Interest was in understanding how value orien-
tations, measured at the psychological level, relate to recy-
cling beliefs and behavior. Because of this interest, we
focused on a limited set of antecedents to the behaviors.
Therefore, several known antecedents to recycling (e.g.,
knowledge, social norms) were not included in the study.
Furthermore, this study was limited to the proenvironmental
behavior of recycling. Thus, although the study has
advanced the understanding of how value orientations relate
to recycling, it does not specifically address recycling in the
broader context of the full range of antecedents to the
behavior, nor does it specifically address the relationship of
these orientations to other environmental behaviors.
A second limitation pertains to population and sample
issues. In our study, the population was limited to commu-
nities where curbside recycling was available. The results
thus may not represent areas where recycling would be
either more difficult (i.e.. no recycling made available by the
municipality) or far more easy (i.e., the separation is done
by the municipality). Clearly, people's perceptions of incon-
venience would be influenced by the ease of recycling in the
area where they live. With respect to the sample, the return
rate was healthy for a mail survey to a general population
sample; however, the nature of the nonrespondents is not
known. Nonresponse bias in surveys is most critical when a
researcher is interested in describing a population rather
than understanding the relationship among a set of variables.
In this respect, the focus of this study was not to report the
incidence of recycling but rather to shed light on the com-
plex relationships between beliefs and behavior. It seems
unlikely that the nature of these relationships would differ
dramatically between people who responded and people
who did not.
It is also worth noting that the effect sizes observed in this
study were not large. At first glance, it might not seem fruit-
ful or cost effective to implement the recommendations we
made with regard to segmentation and targeting. Low effect
sizes are often the norm when measuring input into social
behavior in uncontrolled settings. Our findings simply show
that the variables we measured (however imperfectly) have
an infiuence on a particular behavior, and this influence has
implications for what types of messages might prove most
persuasive. Given the large size of the target (those who
have the ability to recycle) and the relatively low compli-
ance in many cases, even a small change would be welcome.
In this study, we have attempted to provide a better under-
standing for the motivations underlying proenvironmental
behavior. In doing so, we have drawn on theory and
research in several domains, but in particular that which
focuses on abstract but fundamental beliefs about tbe inter-
action of humans with their environment and with other
humans. We have shown that these fundamental psycholog-
ical constructs relate to beliefs and behavior in logical and
predictable ways. Understanding these relations, we believe,
has important implications for understanding and promoting
proenvironmental and other prosocial behaviors.
References
Aaker, Jennifer and Durairaj Maheswaran (1997), "The Effect of
Cultural Orientation on Persuasion," Journal of Consumer
Research, 24 (December), 315-28.
and Patti Williams (1998), "Empathy Versus Pride: The
Influence of Emotional Appeals Across Cultures," Journal of
Consutner Research. 25 (December), 241-61.
 Journal of Public Policy & Marketing
Alwitt, Linda F. and Robeit E. Pitts (1996). "Predicting Purchase
Intentions for an Environmentally Sensitive Product," Journal
of Consumer Psychology, 5 ( 1 ) , 49-64.
Anderson, James C. and David W. Gerbing (1988), "Stnictura!
Equation Modeling in Practice: A Review and Recommended
Two-Step Approach," Psychological Bulletin, 103 (May),
411-23.
Bagozzi, Richard P. and Pratibha A. Dabholkar (1994), "Consumer
Recycling Goals and Their Effect on Decisions to Recycle: A
Means-End Chain Analysis," Psychology and Marketing, 11
(July/August), 3 1 3 ^ 0 .
Balderjahn, Ingo (1988), "Personality Variables and Environmen-
tal Attitudes as Predictors of Ecologically Responsible Con-
sumption Patterns," Journal of Business Research, 17 (August),
51-56.
Baron, Reuben M. and David A. Kenny (1986), 'The Modera-
tor-Mediator Variable Distinction in Social Psychological
Research; Conceptual, Strategic, and Statistical Considera-
tions," Journal of Personality and Social Psychology, 51
(December), 1173-82.
Bellah, Robert N., Richard Madsen, William M. Sullivan, A.
Swindler, and Steven M. Tipton (1985), Habits of the Heart:
Individualism and Commitment in American Life. Berkeley, CA:
University of California Press.
Bentler, Peter M. (1995), EQS Structural Equation Program Man-
ual. Encino, CA: Multivariate Software (nc.
Berger, Ida E. (1997), "The Demographics of Recycling and the
Structure of Environmental Behavior," Environment and Behav-
ior, 29 (Suiy), 5\ 5-31.
and Ruth M. Corbin (1992), "Perceived Consumer Effec-
tiveness and Faith in Others as Moderators of Environmentally
Responsible Behaviors," Journal of Public Policy & Marketing,
11 (Fall), 79-89.
Biswas, Abhijit, Jane W. Licata, Daryl McKee, Chris Pullig, and
Christopher Daughtridge (2(X)0), 'The Recycling Cycle: An
Empirical Examination of Consumer Waste Recycling and
Recycling Shopping Behaviors," Journal of Public Policy &
Marketing, 19 (Spring), 93-105.
Breckler, Steven J., Anthony G. Greenwald, and E.C. Wiggins
(1986). "Public, Private, and Collective Self-Evaluation: Mea-
surement of Individual Differences," paper presented at an
invited international exchange (Poland-U.S.) on self and social
involvement, Princeton, NJ.
Chase, Dennis and Therese Kauchak Smith (1992), "Consumers
Keen on Green but Marketers Don't Deliver," Advertising Age,
63 (June 29), s2-s4.
Dagnoli, Judann (1991), "Consciously Green," Advertising Age, 62
(September 16), 14.
De Young, Raymond (1985-86), "Encouraging Environmentally
Appropriate Behavior The Role of Intrinsic Motivation," Jour-
nal of Environmental Systems, 15 (4), 281-92.
Dietz, Thomas, Paul G. Stem, and Gregory A. Guagnano (1998),
"Social Structural and Social Psychological Bases of Environ-
mental Concern," Environment and Behavior, 30 (July),
450-71.
Dillman, Donald A. (1978), Mail and Telephone Surveys: The
Total Design Method. New York: John Wiley & Sons.
Dunlap, Riley E. and Kent D. Van Liere (1984), "Commitment to
the Dominant Social Paradigm and Concern for Environmental
Quality," Social Science Quarterly, 65 (December), 1013-1028,
103
Gutfeld, Rose (1991), "Eight of 10 Americans Are Environmen-
talists, at Least So They Say," The Wall Street Journal, (August
2), 1.
Han, Sang-Pil (1991), "Individualism and Collectivism: Its Impli-
cations for Cross-Cultural Advertising," doctoral dissertation.
College of Communications, University of Illinois.
and Sharon Shavitt (1994), "Persuasion and Culture:
Advertising Appeals in Individualistic and Collectivistic Soci-
eties," Journal of Experimental Social Psychology, 30 (July),
326-50.
Hines, Jody M., Harold R. Hungerford, and Audrey N. Tomera
(1987), "Analysis and Synthesis of Research on Responsible
Environmental Behavior: A Meta-Analysis," Journal of Envi-
ronmental Education, 18 (Winter), 1-8.
Hofstede, Geert (1980), Culture's Consequences: International
Differences in Work-Related Values. Beveriy Hills, CA: Sage
Publications.
Homer, Pamela and Lynn R. Kahle (1988), "A Structural Equation
Test of the Values-Attitude-Behavior Hierarchy," Journal of
Personality and Social Psychology, 54 (April), 638-46.
Hopper, Joseph R. and Joyce McCad Nielsen (1991), "Recycling
as Altruistic Behavior: Normative Strategies to Expand Partici-
pation in a Community Recycling Program," Environment and
Behavior, 23 (March), 195-220.
Hu, Li-tze and Peter M. Bentler (1998), "Fit Indices in Covariance
Structure Modeling: Sensitivity to Underparameterized Model
Misspecification, Psychological Methods, 3 (December),
424-53.
and (1999), "Cutoff Criteria for Fit Indexes in
Govariance Structure Analysis: Conventional Criteria Versus
New Alternatives," Structural Equation Modeling, 6(1), 1-55.
Hume, Scott and Patricia Stmad (1989), "Consumers Go 'Green,'"
Advertising Age, 60 (September 25), 3, 92.
Jacobs, Harvey E., Jon S. Bailey, and James I. Crews (1984),
"Development and Analysis of a Community-Based Resource
Recovery Progrznx," Journal of Applied Behavior Analysis, 17
(Summer), 121-45.
Kluckhohn, Clyde (1951), "Values and Value Orientations in the
Theory of Action: An Exploration in Definition and Classifica-
tion," in Toward a General Theory of Action, Talcott Parsons
and Edward Shils, eds. Cambridge, MA: Harvard University
Press, 388^33.
Lefcourt, Herbert M. (1991), "Locus of Control," in Measures of
Personality and Social Psychological Attitudes, John P. Robin-
son, Phillip R. Shaver, and Lawrence S. Wrightsman, eds. San
Diego: Academic Press, 413-99.
Leung, Kwok and Michael H. Bond (1984), 'The Impact of Cul-
tural Collectivism on Reward Allocation," Journal of Personal-
ity and Social Psychology, 47 (October), 793-804.
McCarty, John A. and L.J. Shrum (1994), "The Recycling of Solid
Wastes: Personal and Cultural Values and Attitudes About
Recycling as Antecedents of Recycling Behavior," Journal of
Business Research, 30 (May), 53-62.
Paulhus, Delroy (1983), "Sphere-Specific Measures of Perceived
Control," Journal of Personality and Social Psychology, 44
(June), 1253-65.
Pieters, Rik, Tammo Bijmolt, Fred van Raaij, and Mark de Kiuijk
(1998), "Consumers' Attributions of Proenvironmental Behav-
ior, Motivation, and Ability to Self and Others,"
lic Policy & Marketing, 17 (Fall), 215-25.
104 Environmental Beliefs and Behavior
Porter, Bryan E.. Frank C. Leeming, and William O. Dwyer
(1995), "Solid Waste Recovery: A Review of Behavioral Pro-
grams to Increase Recycling," Environment and Behavior, 27
(March). 122-52.
Reitman, Valarie (1992), "Green Product Sales Seem to Be Wilt-
ing." The Wall Street Journal, (May 18), Bl.
Rotter. Julian B. (1966). "Generalized Expectancies for Internal
Versus External Control of Reinforcement," Psychological
Monographs, 80 (1, Whole No. 609).
Schwepker. Charles H. and T. Bettina Comwell (1991), "An
Examination of Ecologically Concerned Consumers and Their
Intention to Purchase Ecologically Packaged Products," Journal
of Public Policy & Marketing, 10 (Fall), 77-101.
Shrum, L.J.. Tina M. Lowrey. and John A. McCarty (1994),
"Recycling as a Marketing Problem: A Framework for Strategy
Development," Psychology and Marketing, 11 (4), 393416.
, John A. McCarty, and Tina M. Lowrey (1995). "Under-
standing the Buyer Characteristics of the Green Consumer:
Implications for Advertising Strategy," Journal of Advertising,
24 (Summer), 71-82.
Sinha, Durganand and Rama Charan Tripathi (1994), "Individual-
ism in a CoUectivistic Culture: A Case of Coexistence of Oppo-
sites," in Individualism and Collectivism: Theory, Method, and
Applications, Uichol Kim et al., eds. Thousand Oaks, CA: Sage
Publications. 123-36.
Sinha, Jai B.P. and Jyoti Verma (1987), "Structure of Collec-
tivism," in Growth and Progress in Cross-Cultural Psychology,
Cigdem Kagitcibasi, ed. Berwyn, PA: Swets North America,
123-29.
Steiger, James H. (1990). "Structural Modei Evaluation and Mod-
ification: An Interval Estimation Approach," Multivariate
Behavioral Research, 25 (April), 173-80.
Thogersen, John (1996). "Recycling and Morality: A Critical
Review of the Literature." Environment and Behavior, 28 (July).
536-58.
Trafimow, David. Harry C. Triandis. and Sharon G. Goto (1991).
"Some Tests of the Distinction Between the Private Self and the
Collective Self," Journal of Personality ami Social Psychology,
60 (May). 649-55.
Triandis, Harry C. (1984), "A Theoretical Framework for the More
Efficient Construction of Cultural Assimilators," International
Journal of Intercultural Relations, 8 (3), 301-30.
(1989), "The Self and Social Behavior in Differing Cul-
tural Contexts," Psychological Review, 96 (July), 506-20.
(1994), "Theoretical and Methodological Approaches to
the Study of Collectivism and Individualism," in Individualism
and Collectivism: Theory, Method, and Applications, Uichol
Kim et al.. eds. Thousand Oaks. CA: Sage Publications, 41-51.
(1995), Individttalism and Collectivism. Boulder, CO:
West view Press.
and Michele J, Gelfand (1998), "Converging Measurement
of Horizontal and Vertical Individualism and Collectivism,"
Journal of Personality and Social Psychology, 74 (January),
118-28.
Turow, Joseph (1997), Breaking Up America: Advertisers and the
New Media World. Chicago: University of Chicago Press.
Vining, Joanne and Angela Ebreo (1990), "What Makes a Recy-
cler? A Comparison of Recyclers and Nonrecyciers." Environ-
ment and Behavior, 22 (January), 55-73.
White, Kathleen M. (1999), "Recycling's Rising Rates," Waste
Age, 30 (May), 194, 200-203.
Williams, Elizabeth (1991), "College Students and Recycling:
Their Attitudes and Behaviors," Journal of College Student
Development, 32 (January). 86-88.